app.py

import streamlit as st
from PIL import Image
import os
import pandas as pd
import tomli

pd.options.mode.chained_assignment = None

st.set_page_config(layout="wide")

# Test change


@st.cache_data
def load_data():
    df_ind = pd.read_csv("data/df_individuals_score.csv", index_col=[0])
    df_ind = df_ind.drop("region_code", axis=1)
    df_ind["productive_year"] = df_ind["productive_year"].astype(int)
    df_ind["individual_wikidata_id"] = "https://www.wikidata.org/wiki/" + df_ind[
        "individual_wikidata_id"
    ].astype(str)

    df_ind = df_ind[df_ind["productive_year"] <= 1800]
    return df_ind


df_ind = load_data()


def load_region_descriptions():
    with open("regions.toml", "rb") as toml_file:
        data = tomli.load(toml_file)
    return data


# Function to get description based on selected region
def get_region_description(region_data, selected_region):
    return region_data[selected_region]["description"]


region_data = load_region_descriptions()


st.sidebar.title("Our History in Data")
st.sidebar.write(
    "This project is led by Charles de Dampierre, Folgert Karsdorp, Mike Kestemont, Valentin Thouzeau and Nicolas Baumard"
)

# Set the global index path
global_index_path = "data/immaterial_index/figures_trends_R/results"
global_index_path_per_capita = (
    "data/immaterial_index/figures_trends_R/results_per_capita"
)
unseen_index_path = (
    "data/immaterial_index/figures_trends_R/figures_unseen/results_unseen"
)
unseen_capita_index_path = (
    "data/immaterial_index/figures_trends_R/figures_unseen/results_unseen/per_capita"
)


population_path = "data/population"
maps_path = "data/map_figures"


from region_list import region_list

region_filtered = list(region_list.keys())

index_paths = {}

for region_key in region_list:
    # Create the index paths for the current region
    index_paths[region_key] = {
        "map": f"{maps_path}/map_{region_key}.png",
        "global_index": f"{global_index_path}/{region_key}.png",
        "global_index_per_capita": f"{global_index_path_per_capita}/{region_key}.png",
        "unseen_index": f"{unseen_index_path}/{region_key}.png",
        "unseen_index_capita": f"{unseen_capita_index_path}/{region_key}.png",
        "population_index": f"{population_path}/{region_key}.png",
    }

# Get the region names (keys) from the index_paths dictionary
regions = list(index_paths.keys())

# Allow the user to select a region
selected_region = st.sidebar.selectbox("Region:", regions, index=regions.index("Japan"))

# Display the selected region's images vertically
if selected_region in index_paths:
    col1, col2 = st.columns(2)

    df = df_ind[df_ind["region_name"] == selected_region]
    df = df.drop(["region_name", "decade"], axis=1)
    df = df[
        [
            "individual_name",
            "productive_year",
            "score",
            "individual_wikidata_id" "",
        ]
    ]
    df = df.sort_values("score", ascending=False)
    df = df.rename(columns={"score": "Number of Catalogs"})

    min_date = region_list[selected_region]["time_range"][0]
    max_date = region_list[selected_region]["time_range"][1]
    df = df[df["productive_year"] >= min_date]
    df = df[df["productive_year"] <= max_date]
    df["productive_year"] = df["productive_year"].astype(int)
    df = df.reset_index(drop=True)

    # Display the data in the left column
    with col1:
        st.header("Cultural Producers")
        st.dataframe(df)
        st.write(f"Number of Cultural producers active before 1800: {len(df)}")

        for key, path in index_paths[selected_region].items():
            if os.path.exists(path):

                if key == "global_index":
                    st.subheader("Global Index")
                    st.image(
                        Image.open(path),
                        caption=key.capitalize(),
                        use_column_width=True,
                    )
                elif key == "global_index_per_capita":
                    st.subheader("Index per capita")
                    st.image(
                        Image.open(path),
                        caption=key.capitalize(),
                        use_column_width=True,
                    )
                elif key == "unseen_index":
                    st.subheader("Unsee-Species Index")
                    st.image(
                        Image.open(path),
                        caption=key.capitalize(),
                        use_column_width=True,
                    )
                elif key == "unseen_index_capita":
                    st.subheader("Unsee-Species per capita Index")
                    st.image(
                        Image.open(path),
                        caption=key.capitalize(),
                        use_column_width=True,
                    )
                elif key == "population_index":
                    st.subheader("Population Index")
                    st.image(
                        Image.open(path),
                        caption=key.capitalize(),
                        use_column_width=True,
                    )
                elif key == "map":
                    st.subheader("Maps")
                    st.sidebar.image(
                        Image.open(path),
                        caption=key.capitalize(),
                        use_column_width=True,
                    )
            else:
                st.write(f"File for {key.capitalize()} does not exist.")

        with col2:
            try:
                region_description = get_region_description(
                    region_data, selected_region
                )
                st.header("Analysis")
                st.write(f"{region_description}")
            except:
                st.write("Analysis not ready yet")